checkpoint-500/trainer_state.json

{
  "best_metric": 0.4143177568912506,
  "best_model_checkpoint": "mikhail-panzo/zlm-fil-ceb_b64_le5_s8000/checkpoint-500",
  "epoch": 21.73913043478261,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.1739130434782608,
      "grad_norm": 0.7357823252677917,
      "learning_rate": 2.5000000000000004e-07,
      "loss": 0.4055,
      "step": 50
    },
    {
      "epoch": 4.3478260869565215,
      "grad_norm": 0.7944665551185608,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.3988,
      "step": 100
    },
    {
      "epoch": 6.521739130434782,
      "grad_norm": 0.7226569652557373,
      "learning_rate": 7.5e-07,
      "loss": 0.4072,
      "step": 150
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 0.7845238447189331,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.4028,
      "step": 200
    },
    {
      "epoch": 10.869565217391305,
      "grad_norm": 0.9769566655158997,
      "learning_rate": 1.25e-06,
      "loss": 0.4066,
      "step": 250
    },
    {
      "epoch": 13.043478260869565,
      "grad_norm": 0.8634074330329895,
      "learning_rate": 1.5e-06,
      "loss": 0.405,
      "step": 300
    },
    {
      "epoch": 15.217391304347826,
      "grad_norm": 0.9029327034950256,
      "learning_rate": 1.75e-06,
      "loss": 0.4074,
      "step": 350
    },
    {
      "epoch": 17.391304347826086,
      "grad_norm": 0.7624074220657349,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.406,
      "step": 400
    },
    {
      "epoch": 19.565217391304348,
      "grad_norm": 1.005196213722229,
      "learning_rate": 2.25e-06,
      "loss": 0.4052,
      "step": 450
    },
    {
      "epoch": 21.73913043478261,
      "grad_norm": 1.0276380777359009,
      "learning_rate": 2.5e-06,
      "loss": 0.4019,
      "step": 500
    },
    {
      "epoch": 21.73913043478261,
      "eval_loss": 0.4143177568912506,
      "eval_runtime": 6.4074,
      "eval_samples_per_second": 24.815,
      "eval_steps_per_second": 3.121,
      "step": 500
    }
  ],
  "logging_steps": 50,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 348,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7395469065410976.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}